Beets Import Guide - What.CD Library¶

Your Library Stats¶

Location: /media/Music/What.CD
Albums: ~2,294 folders
Tracks: 31,391 FLAC files
Estimated Import Time: 10-48 hours (depends on matching speed)

Pre-Import Checklist¶

1. Fix Permissions (Important!)¶

Some folders have permission issues. Let's fix them:

# Shell into Beets
kubectl exec -it -n media deployment/beets -- bash

# Fix permissions on the What.CD folder
# This assumes the NFS share allows changing permissions
chmod -R 755 /media/Music/What.CD 2>/dev/null || echo "Permission fix may need to be done on NFS server"

If that doesn't work, you may need to fix permissions on your NFS server at 10.0.0.2:/mnt/data/media/Music/What.CD

2. Test Import (Highly Recommended)¶

Before importing 2,000+ albums, test with one album:

kubectl exec -it -n media deployment/beets -- bash

# Test with one album
beet import "/media/Music/What.CD/The Shins - Wincing The Night Away"

# This will show you:
# - How Beets matches albums
# - What the organized structure looks like
# - How to answer prompts

Import Strategies¶

Strategy 1: Interactive Import (Most Accurate)¶

Best for: Ensuring perfect matches, reviewing edge cases
Time Required: Days (you need to answer prompts)

kubectl exec -it -n media deployment/beets -- bash

# Install screen to keep session alive
apk add screen

# Start screen session
screen -S beets-import

# Run interactive import
beet import /media/Music/What.CD

# Detach: Ctrl+A then D
# Reattach: screen -r beets-import

Strategy 2: Quiet Mode (Recommended for Large Libraries)¶

Best for: Auto-accepting strong matches, only reviewing uncertain ones
Time Required: 12-48 hours (mostly automated)

kubectl exec -it -n media deployment/beets -- bash
apk add screen
screen -S beets-import

# Quiet mode: auto-accepts matches >90% similarity
beet import -q /media/Music/What.CD

# Detach: Ctrl+A then D

Strategy 3: Incremental Import (Safest)¶

Best for: Testing and gradual migration
Time Required: Weeks (but safest approach)

Import by letter or small batches:

# Import artists starting with 'A'
beet import -q "/media/Music/What.CD/A*"

# Import artists starting with 'B'
beet import -q "/media/Music/What.CD/B*"

# etc...

Strategy 4: Very Quiet Mode (Fastest, Least Accurate)¶

Use only if: You trust your existing metadata
Time Required: 6-12 hours

# Auto-accepts everything, even low similarity
beet import -q -t /media/Music/What.CD

# WARNING: May create incorrect matches!

Recommended Approach for Your Library¶

Given you have What.CD rips (typically well-tagged), I recommend:

Phase 1: Test (Now)¶

kubectl exec -it -n media deployment/beets -- bash

# Test one album
beet import "/media/Music/What.CD/The Shins - Wincing The Night Away"

# Check the result
ls -la /media/music-library/

Phase 2: Small Batch (Today)¶

# Import first 50 albums to verify process
beet import -q "/media/Music/What.CD/[A-C]*"

Phase 3: Full Import (Overnight/Weekend)¶

apk add screen
screen -S beets-import

# Start the full import in quiet mode
beet import -q /media/Music/What.CD > /config/import.log 2>&1

# Detach: Ctrl+A then D

Phase 4: Review (After Import)¶

# Check for albums that couldn't be matched
beet ls -a singleton:true

# Check duplicates
beet duplicates

# View import log
cat /config/import.log

During Import¶

Monitor Progress¶

# Check Beets logs
kubectl logs -n media deployment/beets -f

# Check how many albums imported so far
kubectl exec -n media deployment/beets -- beet stats

# Reattach to screen session
kubectl exec -it -n media deployment/beets -- screen -r beets-import

Import Prompts¶

When Beets finds a match, you'll see:

Artist - Album (2020)
(Similarity: 95.2%)
[A]pply, More candidates, Skip, Use as-is, as Tracks, Enter search, aBort?

Common responses: - A - Apply this match (use this for 90%+ matches) - S - Skip this album - U - Use as-is (keep existing metadata) - M - Show more candidates - I - Enter MusicBrainz ID manually

Expected Timeline¶

Quiet mode (-q) estimate: - ~10-20 seconds per album (on average) - 2,294 albums × 15 seconds = ~9.5 hours - Real time: 12-24 hours (accounting for network, MusicBrainz API limits)

After Import¶

1. Verify Organization¶

kubectl exec -n media deployment/beets -- bash

# Check organized library
ls -la /media/music-library/

# Get stats
beet stats

# Sample output:
# Tracks: 31391
# Total time: 98.5 days
# Approximate total size: 500 GB
# Artists: 1200
# Albums: 2294

2. Update Plex¶

Plex Settings → Libraries → Music
Edit Library or Add Library:
Folder: /media/music-library
Scanner: Plex Music
Agent: Plex Music
✅ Use embedded tags
Scan Library

3. Keep Original Files (Optional)¶

After verifying the organized library works in Plex:

# Your original files are still at:
/media/Music/What.CD

# Beets MOVED them to:
/media/music-library/Artist/Album/Tracks.flac

# The What.CD folder should now be empty or nearly empty
# (Some files may remain if they couldn't be matched)

4. Handle Unmatched Files¶

# Find albums that couldn't be matched
beet ls -a singleton:true

# These are stored in:
/media/music-library/Non-Album/

# You can:
# - Manually import them later
# - Fix tags and re-import
# - Leave them as-is

Troubleshooting¶

Permission Denied Errors¶

# Fix from NFS server at 10.0.0.2
ssh admin@10.0.0.2
chmod -R 755 /mnt/data/media/Music/What.CD
chown -R 568:568 /mnt/data/media/Music/What.CD

Slow Import Speed¶

# MusicBrainz API rate limiting is normal
# You can't speed this up much, but you can:
# - Use -q flag (reduces prompts)
# - Import in smaller batches
# - Be patient (it's worth it!)

Failed Matches¶

# If many albums fail to match:
# 1. Check internet connectivity
# 2. Verify MusicBrainz is accessible
# 3. Try manual search with 'I' option
# 4. Use 'U' to keep existing metadata

Out of Disk Space¶

# Check available space
kubectl exec -n media deployment/beets -- df -h /media

# Beets MOVES files, so total space used should be the same
# If you're low on space, import in batches

Commands Reference¶

# Access Beets container
kubectl exec -it -n media deployment/beets -- bash

# Start import
beet import -q /media/Music/What.CD

# Check progress
beet stats

# View logs
cat /config/beet.log

# List imported albums
beet ls -a

# Find duplicates
beet duplicates

# Update metadata
beet update

# Fetch missing art
beet fetchart

Next Steps¶

Run test import on one album
Verify the organized output in /media/music-library
Start small batch (50-100 albums)
Check in Plex to make sure it looks good
Start full import in screen session
Wait 12-48 hours for completion
Update Plex to use /media/music-library
Enjoy your perfectly organized library!

Ready to start? Run this:

kubectl exec -it -n media deployment/beets -- bash

Then paste:

# Test with one album first
beet import "/media/Music/What.CD/The Shins - Wincing The Night Away"